Exploring the reactivity of carbon(0)/borane-based frustrated lewis pairs

Article abstract of DOI:10.1002/anie.201002119

chemical equation presented Twice frustrated: The unusual electronic distribution around the central carbon(0) atom in carbodiphosphoranes makes this center so basic that, even after a first alkylation step, it is still able to act as a cationic Lewis bas

Full text of DOI:10.1002/anie.201002119

Communications
DOI: 10.1002/anie.201002119
Frustrated Lewis Pairs
Exploring the Reactivity of Carbon(0)/Borane-Based Frustrated Lewis
Pairs**
Manuel Alcarazo,* Catherine Gomez, Sigrid Holle, and Richard Goddard
Dedicated to Professor Rosario Fernꢀndez
Since the unveiling of the concept of frustrated Lewis pairs
(FLPs) by Stephan et al.,[1] the chemistry of these systems has
flourished. Arguably their most attractive application has
been the heterolytic activation of H2,[2] and the subsequent
development of metal-free hydrogenation catalysis directly
employing dihydrogen[3] rather than a surrogate.[4] Although
Initially, we carried out the reaction of 1 with B(C6F5)3 in
toluene at room temperature. The NMR spectroscopic data of
the obtained product suggested formation of 2 by nucleophilic
attack at the para position of the pentafluorophenyl ring and
trapping of the generated fluoride anion by the boron atom.
X-ray crystallographic analysis later confirmed the structure
of 2 (see Supporting Information).[18] However, when the
same reagents were mixed at ꢀ788C, NMR spectroscopy
indicated no interaction between the partners, that is,
“frustration”. Purging this stoichiometric mixture with H2
resulted in formation of a white precipitate which could be
isolated in 91% yield. The NMR data support the formulation
for this product as [(Ph3P)2CH][HB(C6F5)3] (3; Scheme 1).
The cation exhibits a 1H signal at d = 1.73 ppm with 2J(1H,31P)
of 5.4 Hz and a 31P{1H} resonance at d = 21.3 ppm, while the
[5]
[6]
[7]
several other bonds such as C O, C H, B H, S S,[8] and
ꢀ
ꢀ
ꢀ
ꢀ
[9]
ꢀ
C C, have also been cleaved by using FLPs, these systems
largely rely on P- or N-based Lewis bases combined with a
polyfluorinated borane.[10] The sole exceptions are the steri-
cally crowded carbene 1,3-di-tert-butyl-1,3-imidazol-2-ylidene
(ItBu) in combination with B(C6F5)3, a pair that contains a C-
derived base,[11] and the use of Al(C6F5)3 instead of a borane
as Lewis acid.[12] Clearly, extension of the FLP concept to
include currently unexplored partners is desirable, as it may
lead to the discovery of a range of interesting new applica-
tions.
In our research to broaden the range of bases that can be
used in FLP chemistry, we were inspired by the computational
investigations of Tonner and Frenking on the nature of
carbodiphosphoranes.[13,14] They proposed that these com-
pounds should be considered to comprise two phosphine
ligands coordinated to a central zero-valent carbon atom that
retains its four valence electrons. This view has been
subsequently confirmed experimentally by the work of
Bertrand et al., Fꢀrstner et al., and others.[15]
The available information suggests that C0 compounds
must be exceptionally good nucleophiles. In fact, the calcu-
lated proton affinity for carbodiphosphoranes surpasses the
values reported for amines, phosphines, and even N-hetero-
cyclic carbenes. It can be envisaged therefore that, if
sufficiently sterically hindered, C0 compounds should be
qualified to function as bases in the framework of FLP
chemistry. Herein, in an attempt to address this issue, the pair
hexaphenylcarbodiphosphorane (1)/B(C6F5)3 is studied and
its reactivity towards several small molecules evaluated.[17]
Scheme 1. Some reactions of FLP 1/B(C6F5)3. Reagents and conditions
(yields): a) toluene, ꢀ788C!RT (74%); b) H2 1 atm, toluene,
ꢀ788C!RT (91%); c) THF, ꢀ788C!RT (76%); d) nC5H11F, toluene,
ꢀ788C, quant.; e) ethylene carbonate, toluene, ꢀ788C!RT (84%);
f) 2,2-dimethyl-g-butyrolactone, toluene, 788C!RT (71%).
anion gives the expected 11B signal of a borohydride at d =
1
ꢀ25.5 ppm with a J(1H,11B) of 100 Hz. Single crystals were
obtained by slow diffusion of pentane into a solution of 3 in
CH2Cl2, and X-ray structure analysis confirmed not only the
proposed structure (Figure 1), but also the ability of 1/
B(C6F5)3 to function as an FLP.
[*] Dr. M. Alcarazo, Dr. C. Gomez, S. Holle, Dr. R. Goddard
Max Planck Institut fꢀr Kohlenforschung
45470 Mꢀlheim and der Ruhr (Germany)
Fax: (+49)208-306-2994
As expected for an FLP, 1/B(C6F5)3 in solution in THF
resulted in ring opening of the ether to produce phosphonio
borate 4, also confirmed by X-ray analysis (see Supporting
Information). This reactivity was extended to ethylene
carbonate and the non-enolizable ester 3,3-dimethyl-g-butyr-
olactone to produce zwitterionic species 5 and 6, respectively.
Moreover, addition of one equivalent of 1-fluoropentane to a
suspension of 1/B(C6F5)3 at ꢀ788C generates [(Ph3P)2C-
E-mail: alcarazo@mpi-muelheim.mpg.de
[**] We thank Prof. A. Fꢀrstner for generous support and constant
encouragement. The NMR spectroscopy and X-ray crystallography
departments of our institute are also gratefully acknowledged for
excellent support, as well as H. Bruns and R. Schinzel for the
preparation of starting materials.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002119.
5788
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5788 –5791

Angewandte
Chemie
Figure 2. Molecular structure of 9·CH2Cl2 in the solid state (hydrogen
atoms removed for clarity; ellipsoids set at 50% probability).[16]
Figure 1. Molecular structure of 3 in the solid state (except for the
hydrogen atom bonded to B, hydrogen atoms and solvent molecules
were removed for clarity; ellipsoids set at 50% probability).[16]
ꢀ
(C5H11)][FB(C6F5)3] in quantitative yield. This C F bond
activation of an alkyl fluoride is unprecedented in the scope of
FLP chemistry and does not take place in the absence of
borane, even at room temperature.[19]
The reactivity of terminal alkynes such as phenylacetylene
in presence of several P/B- and P/Al-based FLPs has been
studied by Stephan and co-workers.[6] Two possible reaction
pathways, involving either deprotonation of the alkyne or
Scheme 2. Reactivity of the pair 1/B(C6F5)3 towards terminal alkynes.
Reagents and conditions (yields): a) phenylacetylene, toluene
ꢀ788C!RT, 8 (78%), 9 (12%); b) Ph2SiH2, toluene, ꢀ788C!RT
(87%); c) Me2EtSiH, toluene, ꢀ788C!RT (93%); d) Me3SiOMe,
toluene, ꢀ788C!RT, hydrolysis (88%).
ꢀ
Markovnikov addition of the FLP to the C C triple bond,
were observed. This divergent reactivity was explained in
terms of the Brønsted basicity of the phosphines used. The
ꢀ
very basic tBu3P prompts deprotonation of the alkyne C H,
while aryl phosphines, which are less basic, prefer nucleo-
philic attack on the activated alkyne. Nonetheless, this
rationale does not explain the fact that treatment of phenyl-
acetylene with 1/B(C6F5)3, whereby 1 is a stronger base than
tBu3P, resulted in formation of both possible products, that is,
the expected 8 but also 9. This points towards a concomitant,
previously unconsidered, influence of steric factors. The X-
ray structure of 9 is shown in Figure 2.
with a chiral silane.[21] Likewise, an equimolar solution of 1
and Ph2SiH2 in benzene does not show any reactivity even
after a day at room temperature. These results confirmed that
the synergic effect of the base and the acid of our FLP is
ꢀ
responsible for activation of the Si H bonds. To the best of
ꢀ
our knowledge, the achieved activation of Si H bonds also
ꢀ
Activation of Si H bonds with the FLP 1/B(C6F5)3 was
has no precedent in FLP chemistry.
ꢀ
ꢀ
also attempted. Upon stirring an equimolar mixture of
Ph2SiH2, 1 and B(C6F5)3 at ꢀ788C in toluene, the yellow
color of the suspension slowly vanished and an immiscible
colorless oil separated from the toluene phase. Purification by
column chromatography afforded a pure compound that
In contrast, heterolytic cleavage of Si F and Si O bonds
in PhMe2SiF and Me3SiOPh was not successful under the
same reaction conditions, and only 2 was obtained after
allowing the reaction mixture to reach room temperature.
When Me3SiOMe was tested, product 12 was isolated. This
1
3
showed H signals at d = 5.12 ppm with J(1H,31P) of 16 Hz
compound presumably results from activation of the C O
ꢀ
and at d = 3.54 ppm with J(1H,11B) of 92.6 Hz, which can be
bond of the silyl ether and hydrolysis of the trimethylsilyl
borate during chromatographic purification.
1
assigned to a silane proton and a borohydride proton
respectively. A 31P{1H} resonance at d = 26.7 ppm and a
11B{1H} signal at d = ꢀ23.8 ppm, together with high-resolution
mass spectra of both the anion and the cation, unambiguously
confirmed the proposed structure for 10 (Scheme 2).[20]
Frenking estimated the second proton affinity of 1 to be
ꢀ193.4 kcalmolꢀ1. This basicity is considered a hallmark of
carbon(0)-containing compounds and suggests that even
protonated or alkylated derivatives of 1 may still have an
appreciable degree of frustration when confronted with
B(C6F5)3. To explore this possibility, salt 13 consisting of the
methylated cation [1·Me]+ and a noninterfering [B(C6F5)4]ꢀ
anion was synthesized.[17] Although the pair 13/B(C6F5)3 was
ꢀ
Activation of the Si H bond in EtMe2SiH was also achieved
under the same reaction conditions to give 11. Formation of a
free silylium cation by direct interaction of the silane with
B(C6F5)3 has been ruled out by the work of Oestreich and
Render on the enantioselective reduction of acetophenone
ꢀ
not able to activate H2 or N H of amines, it still cleaves the
Angew. Chem. Int. Ed. 2010, 49, 5788 –5791
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5789

Communications
ꢀ
O H bond of methanol, something that 13 alone is not able to
do (Scheme 3). In fact, 13 can be crystallized from MeOH
frustration persists. The application of these FLPs in homo-
geneous catalysis is currently under investigation.
without generating any protonated product.
Received: April 9, 2010
Published online: July 2, 2010
Keywords: boranes · carbodiphosphoranes · donor–
.
acceptor systems · frustrated Lewis pairs · hydrogen activation
ꢀ
Scheme 3. Cleavage of the O H bond of methanol with the pair 13/
B(C6F5)3. Reagents and conditions (yields): a) MeOH (1 equiv),
toluene, RT, 14 (86%).
[1] G. C. Welch, R. R. S. Juan, J. D. Masuda, D. W. Stephan, Science
2006, 314, 1124 – 1126. For some reviews on the chemistry of
frustrated Lewis pairs, see: a) D. W. Stephan, G. Erker, Angew.
Chem. 2010, 122, 50 – 81; Angew. Chem. Int. Ed. 2010, 49, 46 – 76;
b) D. W. Stephan, Org. Biomol. Chem. 2008, 6, 1535 – 1539.
[2] For theoretical mechanistic studies, see: a) T. A. Rokob, A.
Hamza, A. Stirling, T. Soꢁs, I. Pꢂpai, Angew. Chem. 2008, 120,
2469 – 2472; Angew. Chem. Int. Ed. 2008, 47, 2435 – 2438;
b) T. A. Rokob, A. Hamza, I. Pꢂpai, J. Am. Chem. Soc. 2009,
131, 10701 – 10710; c) S. Grimme, H. Kruse, L. Goerigk, G.
Erker, Angew. Chem. 2010, 122, 1444 – 1447; Angew. Chem. Int.
Ed. 2010, 49, 1402 – 1405; d) For splitting of dihydrogen by using
a single carbon center, see: G. D. Frey, V. Lavallo, B. Donnadieu,
W. W. Schoeller, G. Bertrand, Science 2007, 316, 439 – 441.
[3] a) P. Spies, S. Schwendermann, S. Lange, G. Kehr, R. Frꢃlich, G.
Erker, Angew. Chem. 2008, 120, 7654 – 7657; Angew. Chem. Int.
Ed. 2008, 47, 7543 – 7546; b) P. A. Chase, G. C. Welch, T. Jurca,
D. W. Stephan, Angew. Chem. 2007, 119, 8196 – 8199; Angew.
Chem. Int. Ed. 2007, 46, 8050 – 8053; c) H. Wang, R. Frꢃhlich, G.
Kehr, G. Erker, Chem. Commun. 2008, 5966 – 5968; d) T. A.
Rokob, A. Hamza, A. Stirling, I. Pꢂpai, J. Am. Chem. Soc. 2009,
131, 2029 – 2036.
This reveals a degree of frustration that, albeit clearly
diminished in comparison to the 1/B(C6F5)3 pair, is still
remarkable considering the cationic nature of the employed
base. Cationic Lewis acids like [CPh3]+ have been studied in
the context of FLPs;[12] however, this is the first example of
the counterintuitive use of a cationic base in this field. The
formulation of compound 14 is supported by a 1H NMR signal
3
at d = 1.95 ppm for the Me group with J(1H,31P) of 16.0 Hz
and a 3J(1H,1H) of 7.6 Hz, a 31P{1H} resonance at d = 28.6 ppm
and two 11B{1H} signals at d = ꢀ2.3 and ꢀ15.2 ppm. In
addition, during attempts to crystallize 14, crystals of 14[B-
(C6F5)4]2 were obtained and its structure confirmed by X-ray
analysis (Figure 3).
[4] a) J. W. Yang, M. T. Hechavarria Fonseca, B. List, Angew. Chem.
2004, 116, 6829 – 6832; Angew. Chem. Int. Ed. 2004, 43, 6660 –
8053; b) J. B. Tuttle, S. G. Ouellet, D. W. C. MacMillan, J. Am.
Chem. Soc. 2006, 128, 12662 – 12663.
[5] a) S. J. Geier, D. W. Stephan, J. Am. Chem. Soc. 2009, 131, 3476 –
3477; b) G. C. Welch, J. D. Masuda, D. W. Stephan, Inorg. Chem.
2006, 45, 478 – 480.
[6] M. A. Dureen, D. W. Stephan, J. Am. Chem. Soc. 2009, 131,
8396 – 8397.
[7] M. A. Dureen, A. Lough, T. M. Gilbert, D. W. Stephan, Chem.
Commun. 2008, 4303 – 4305.
[8] M. A. Dureen, G. C. Welch, T. M. Gilbert, D. W. Stephan, Inorg.
Chem. 2009, 48, 9910 – 9917.
[9] a) J. S. J. McCahill, G. C. Welch, D. W. Stephan, Angew. Chem.
2007, 119, 5056 – 5059; Angew. Chem. Int. Ed. 2007, 46, 4968 –
4971; b) M. Ullrich, K. S. H. Seto, A. J. Lough, D. W. Stephan,
Chem. Commun. 2009, 2335 – 2337; C. M. Mꢃmming, G. Kehr, B.
Wibbeling, R. Frꢃhlich, B. Schirmer, S. Grimme, G. Erker,
Angew. Chem. 2010, 122, 2464 – 2467; Angew. Chem. Int. Ed.
2010, 49, 2414 – 2417.
[10] a) P. Spies, G. Erker, G. Kehr, K. Bergander, R. Frꢃlich, S.
Grimme, D. W. Stephan, Chem. Commun. 2007, 5072 – 5074;
b) C. M. Mꢃmming, S. Frꢃmel, G. Kehr, R. Frꢃhlich, S. Grimme,
G. Erker, J. Am. Chem. Soc. 2009, 131, 12280 – 12289.
[11] a) D. Holschumacher, T. Bannenberg, C. G. Hrib, P. G. Jones, M.
Tamm, Angew. Chem. 2008, 120, 7538 – 7542; Angew. Chem. Int.
Ed. 2008, 47, 7428 – 7432; b) P. A. Chase, D. W. Stephan, Angew.
Chem. 2008, 120, 7543 – 7547; Angew. Chem. Int. Ed. 2008, 47,
7433 – 7437.
Figure 3. Molecular structure of 14[B(C6F5)4]2 in the solid state.
(B(C6F5)4 anions and hydrogen atoms, except that bonded to C1, were
removed for clarity; ellipsoids set at 50% probability).[16]
In conclusion, this study demonstrates that the combina-
tion of hexaphenylcarbodiphosphorane as carbon-based
Lewis base and B(C6F5)3 forms a novel frustrated Lewis
ꢀ
pair. The system is not only able to achieve the classical H H,
ꢀ
ꢀ
C O, and C H bond cleavages, but also unprecedented
[12] The use of carbon-based Lewis acids for FLP chemistry has been
tried but with limited success: L. Cabrera, G. C. Welch, J. D.
Masuda, P. Wei, D. W. Stephan, Inorg. Chim. Acta 2006, 359,
3066 – 3071.
ꢀ
activation of Si H bonds and alkyl fluorides. The peculiar
electronic situation of the C0 base makes the system unique:
after the first protonation or alkylation, some degree of
5790 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 5788 –5791

Angewandte
Chemie
[13] a) R. Tonner, G. Frenking, Chem. Eur. J. 2008, 14, 3260 – 3272;
b) R. Tonner, G. Frenking, Chem. Eur. J. 2008, 14, 3273 – 3289;
c) R. Tonner, G. Frenking, Angew. Chem. 2007, 119, 8850 – 8853;
Angew. Chem. Int. Ed. 2007, 46, 8695 – 8698; d) R. Tonner, F.
ꢄxler, B. Neumꢀller, W. Petz, G. Frenking, Angew. Chem. 2006,
118, 8206 – 8211; Angew. Chem. Int. Ed. 2006, 45, 8038 – 8042.
See also W. C. Kaska, D. K. Mitchell, R. F. Reichelderfer, J.
Organomet. Chem. 1973, 47, 391 – 402.
supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
[17] F. Ramꢅrez, N. B. Desai, B. Hansen, N. McKelvie, J. Am. Chem.
Soc. 1961, 83, 3539 – 3540. See also a) N. D. Jones, R. G. Cavell, J.
Organomet. Chem. 2005, 690, 5485 – 5496; b) W. C. Kaska, D. K.
Mitswchell, R. F. Reichelerfer, J. Organomet. Chem. 1973, 47,
391 – 402.
[14] For a review on the coordination chemistry of ylides and
carbodiphosphoranes, see: H. Schmidbaur, Angew. Chem. 1983,
95, 980 – 1000; Angew. Chem. Int. Ed. Engl. 1983, 22, 907 – 927.
[15] a) C. A. Dyker, V. Lavallo, B. Donnadieu, G. Bertrand, Angew.
Chem. 2008, 120, 3250 – 3253; Angew. Chem. Int. Ed. 2008, 47,
3206 – 3209; b) M. Alcarazo, C. W. Lehmann, A. Anoop, W.
Thiel, A. Fꢀrstner, Nat. Chem. 2009, 1, 294 – 301; c) C. A. Dyker,
G. Bertrand, Nat. Chem. 2009, 1, 265 – 266; d) A. Fꢀrstner, M.
Alcarazo, R. Goddard, C. W. Lehmann, Angew. Chem. 2008,
120, 3254 – 3258; Angew. Chem. Int. Ed. 2008, 47, 3210 – 3214;
e) M. Melaimi, P. Parameswaran, B. Donnadieu, G. Frenking, G.
Bertrand, Angew. Chem. 2009, 121, 4886 – 4889; Angew. Chem.
Int. Ed. 2009, 48, 4792 – 4795; f) J. Vicente, A. R. Singhal, P. G.
Jones, Organometallics 2002, 21, 5887 – 5900.
[18] This observed reactivity has a precedent in the case of some
phosphines and bulky phosphorus ylides: a) G. C. Welch, R.
Prieto, M. A. Dureen, A. J. Lough, O. A. Labeodan, T. Hꢃll-
trichter-Rꢃssmann, D. W. Stepan, Dalton Trans. 2009, 1559 –
1570; b) S. Dꢃring, G. Erker, R. Frꢃhlich, O. Meyer, K.
Bergander, Organometallics 1998, 17, 2183 – 2187.
[19] Alkylations of carbodiphosphorane 1 are known but only if
harsher reaction conditions and alkyl chlorides or bromides are
employed: H. J. Bestmann, H. Oechner, Z. Naturforsch. B 1983,
38, 861 – 865.
[20] Preliminary studies show that the well-established tBu3P/B-
(C6F5)3 pair under the same conditions produces [tBu3PH]
ꢀ
[HB(C6F5)3], presumably by hydrolysis of the Si P bond during
the workup or purification.
[16] CCDC 772576 (2), 772577 (3), 772578 (4), 772579 (5), 772580 (6),
772581 (8), 772582 (9), 772583 (13) and 772593 (14) contain the
[21] S. Rendler, M. Oestreich, Angew. Chem. 2008, 120, 6086 – 6089;
Angew. Chem. Int. Ed. 2008, 47, 5997 – 6000.
Angew. Chem. Int. Ed. 2010, 49, 5788 –5791
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5791